analysis Exploring Optimal Prices for 4L .com Domains

[4L.com]

Hello everyone,

As some members of this forum may know, I have an extensive portfolio of 4L com domains. Starting from October 1, 2024, I’ve decided to embark on a series of experiments to determine the most profitable pricing strategy for selling these domains. Currently, I own over 16,000 4L domains that contain one or more of the letters J, Q, U, V, W, X, Y, or Z. These are the domains I’ll be using for my pricing experiments. I plan to continue buying domains, so the number will gradually increase.

I aim to find out which approach will be more profitable: selling fewer domains at a higher price or selling more domains at a lower price. Initially, I wanted to start this experiment on September 1, but due to the issues with Afternic, which compromised the accuracy of any data, I decided to wait until October when most of these issues were resolved.

At the moment, I’ve set the price for all my domains at $14,999 each. The nameservers are configured to ns1.afternic.com and ns2.afternic.com. I’ve disabled Boost as I don’t find it useful. I’ve chosen the "Custom Lander" option, and all domains have a Lease-to-Own (LTO) option available for a 12-month term. The "Make Offer" option is disabled, and I will be ignoring any inquiries through Whois during the experiment to ensure that potential buyers can only purchase the domains at a fixed price without the opportunity to negotiate discounts or terms. This also prevents them from contacting me directly, as well as avoiding back-and-forth with brokers who often try to negotiate regardless of the set price. The same applies to Afternic brokers - I will reject any offers or discounts they send through the dashboard unless the domain is purchased at the current fixed price.

My goal is to achieve the best possible pricing strategy that minimizes the time I spend managing this process. Therefore, there will be no brokers, no email negotiations, no personal landing pages - just a fixed price on Afternic.

The planned schedule for the experiment is as follows:

• From October 1, 2024, to November 30, 2024, the fixed price for all domains will be $14,999.
• From December 1, 2024, to January 31, 2025, the fixed price will be $9,999.
• From February 1, 2025, to March 31, 2025, the fixed price will be $6,999.
• From April 1, 2025, to May 31, 2025, the fixed price will be $4,999.
• From June 1, 2025, to July 31, 2025, the fixed price will be $3,999.
• From August 1, 2025, to September 30, 2025, the fixed price will be $2,999.
• From October 1, 2025, to November 30, 2025, the fixed price will be $1,999.
• From December 1, 2025, to January 31, 2026, the fixed price will be $999.

I believe that two months per pricing stage and a pool of >16,000 domains should provide enough data to draw meaningful conclusions for future decision-making.

I plan to post monthly updates on the results in this thread, assuming it doesn’t violate any forum rules. I know that many forum members actively invest in 4L domains containing these letters (J, Q, U, V, W, X, Y, Z), so I believe the results could be valuable for others as well. I haven’t come across any detailed statistics on pricing for these domains, which is why I’ve decided to conduct this research myself.

If anyone has any insights or ideas to make this experiment more beneficial, both for myself and for other forum members, I’d be happy to hear them.

 

[davidc1]

Imo the test is not good.
It would be better to divide your domains randomly, split testing each bucket with one price. Waiting maybe for maybe 100 sales, then keep one or two or three winners bucket price depending of the results. Then after more time, keep the best one for all domains.

 

[4L.com]

Imo the test is not good.
It would be better to divide your domains randomly, split testing each bucket with one price. Waiting maybe for maybe 100 sales, then keep one or two or three winners bucket price depending of the results. Then after more time, keep the best one for all domains.

Hello, thank you for your input. I agree that this approach might be more effective, but the downside is that it would take a very long time to gather reliable results for the higher-priced domains. If we divide 16,000 domains by 8 price tiers, that gives us 2,000 domains per price range. While domains priced at $999 or $1,999 might see 1-4 sales per month, the higher-priced domains could go months without any sales, and domains priced at $14,999 or $9,999 might see only 1 or 2 sales over the course of a year, if any at all.

An alternative approach could be to reduce the number of price ranges to four and start testing with lower prices. For instance, split the 16,000 domains into four groups:

1. 4,000 domains at $999
2. 4,000 domains at $1,999
3. 4,000 domains at $2,999
4. 4,000 domains at $3,999

Wait for 6-8 months and then review the results.

Alternatively, split the domains into three groups:

1. 5,333 domains priced at $999
2. 5,333 domains priced at $1,999
3. 5,333 domains priced at $2,999

This might give more balanced data over a shorter period of time.

 

[davidc1]

Choose properly the number of group, you are true...

The problem if you lower price month after month, you'll see that you'll make much less sales if you put higher prices the second year as many have been sold for less the first year....

 

[4L.com]

I still want to test the price of 14,999 USD over a period of 2 months to get at least some results that apply to my domains. Logically, 16,000*10.5 = 168,000 just for renewal costs, and that’s 168,000/12 = 14,000 USD per month. I’d like to believe that at least a couple of domains will sell over these 2 months to cover the renewal expenses

After that, I can run a longer experiment by splitting the domains into 3-4 pricing groups with lower prices to see the results, as they should come faster compared to higher-priced domains.

 

[davidc1]

I still want to test the price of 14,999 USD over a period of 2 months to get at least some results that apply to my domains. Logically, 16,000*10.5 = 168,000 just for renewal costs, and that’s 168,000/12 = 14,000 USD per month. I’d like to believe that at least a couple of domains will sell over these 2 months to cover the renewal expenses

After that, I can run a longer experiment by splitting the domains into 3-4 pricing groups with lower prices to see the results, as they should come faster compared to higher-priced domains.

Good idea

 

[4L.com]

The problem if you lower price month after month, you'll see that you'll make much less sales if you put higher prices the second year as many have been sold for less the first year....

My assumption is based on the idea that there is a constant demand for 4L domains, and the previous price history of the domains is irrelevant. Each day, for example, there are around 10 hypothetical potential buyers who are looking for unique domain letter combinations. They discover my domains and make a decision that day whether to purchase at the current price or select another domain. The next day, 10 more, and so on. Of course, some percentage of them may add the domain to their favorites and monitor the price, but I don’t believe that percentage is very high.

 

[davidc1]

My assumption is based on the idea that there is a constant demand for 4L domains, and the previous price history of the domains is irrelevant. Each day, for example, there are around 10 hypothetical potential buyers who are looking for unique domain letter combinations. They discover my domains and make a decision that day whether to purchase at the current price or select another domain. The next day, 10 more, and so on. Of course, some percentage of them may add the domain to their favorites and monitor the price, but I don’t believe that percentage is very high.

Ok i see. But dont you think a part of buyers are checking price sometimes, 1-3 times per year for the same domain and if the price is ok they buy it.

 

[DN_Hunter]

Hello and thank you for conducting this research!

Did you apply any filtering to exclude "better" letter combinations (even with J,Q,U,V,W,X,Y,Z)? Such as names that
1) begin or end with good abbreviations (MY, VR, HQ, FX, etc...)
2) names that contain a 3-letter word + 1 random letter
3) letter patterns (CVCV, VCVC, etc...)
4) very pronounceable
5) potentially could be used as a person's name

Also, at $14,999 would you consider a longer lease period? Perhaps up to 24 or 30 months. Makes it a little easier for the buyer to pay over time, with the added bonus of lowering your Afternic commission rate.

Good luck with this experiment, and I look forward to the results!

 

[789.xyz]

What is the purpose of your testing? A domain name is not like Bitcoin; it is unique.

Even if AAAA.com sells for $14,999, it does not mean that all LLLL-type domain names are worth $14,999.

 

[korganian]

I don't think that is a good idea. You give potential buyers every reason to wait for the price to keep dropping if they had previously checked the price. Then once you reach $999, that is the price buyers will feel entitled to buy your names for from that point forward.

 

[4L.com]

Did you apply any filtering to exclude "better" letter combinations (even with J,Q,U,V,W,X,Y,Z)? Such as names that
1) begin or end with good abbreviations (MY, VR, HQ, FX, etc...)
2) names that contain a 3-letter word + 1 random letter
3) letter patterns (CVCV, VCVC, etc...)
4) very pronounceable
5) potentially could be used as a person's name

The domains from points 1-5 are definitely not on this list Just random letters.

Also, at $14,999 would you consider a longer lease period? Perhaps up to 24 or 30 months. Makes it a little easier for the buyer to pay over time, with the added bonus of lowering your Afternic commission rate.

Regarding longer periods, I have had a similar experience with Dan. For domains with a high price and a 24-60 month payment period, a buyer would come along, pay for the first month, and then stop paying. It's obvious that they needed the domain for some temporary purposes (to get an email from/to this exact domain, conduct some scam, try to resell it, etc.). And this didn’t happen just once or twice - it happened regularly, so I decided to stop offering LTO periods longer than 12 months for all of my domains.

What is the purpose of your testing? A domain name is not like Bitcoin; it is unique.

As I mentioned at the very beginning, I want to find the price that will bring me the maximum benefit. My minimum requirement is that the income from selling the domains, after deducting all expenses, covers the renewal costs. If I can also make some profit on top of that, even better.

I don't think that is a good idea. You give potential buyers every reason to wait for the price to keep dropping if they had previously checked the price. Then once you reach $999, that is the price buyers will feel entitled to buy your names for from that point forward.

It will only be a small percentage of people who are persistent enough to keep track. I've changed prices several times from around 10,000 to 2,000, and I can say for sure - there are very few such people. There is no surge in sales, which means almost no one is tracking the domain's price. Again, this is specifically about my domains (4L with random letters), and perhaps the situation is different for other types of domains.

 

[clarkemarketing]

This experiment sounds awesome, I own a bunch of 4L.com and love them so I’m very excited for your updates.

 

[AEProgram]

since you made up your mind to accept 999, save time by listing all at 14999 with a 999 floor. The only way you should try your way is if you make your own lander and have people enter their email to get an instant email with the price. This way you collect emails and when you lower the price you have who to email. You will do the best with this setup.

 

[davidc1]

since you made up your mind to accept 999, save time by listing all at 14999 with a 999 floor. The only way you should try your way is if you make your own lander and have people enter their email to get an instant email with the price. This way you collect emails and when you lower the price you have who to email. You will do the best with this setup.

I tried that in the past, it doesnt worked well.

 

[papeter]

I own BUNH.com, do you think I have any chance to sell it at $1999?

 

[mdent]

Thanks for doing this! As a fellow 4 letter enthusiast, this data could be super valuable.

An alternative setup to consider:

Instead of doing 8 tiers over time, run 3 randomized pricing tiers concurrently. You'll still gather data relatively fast and it will be less affected by trends, buyers monitoring the price, and seasonal changes like most startups launching in the spring, etc.
Start with something like 10k, 6k, 3k. After 6 months, you could expand the best performing tier(s) to fine tune the pricing. Let's say the 6k tier produced the best results by far, great, re-randomize the tiers to 8k, 7k, 6k, 5k, 4k and let them run for the rest of the experiment.

If you need help with anything, drop me a DM

 

[byte]

Do you have a link to your portfolio?

 

[Omit]

interesting

I share my experience. I have a very small domain portfolio, and I offer a 25-35% discount twice a year: the first is a summer discount from July 20 to September 10, and the second is a Christmas promo from November 20 to January 10. As a result, I sold 4 domains during these discount periods (out of a total of 6 sales for my 4L.com domains).”

 

[4L.com]

Thanks for doing this! As a fellow 4 letter enthusiast, this data could be super valuable.

An alternative setup to consider:

Instead of doing 8 tiers over time, run 3 randomized pricing tiers concurrently. You'll still gather data relatively fast and it will be less affected by trends, buyers monitoring the price, and seasonal changes like most startups launching in the spring, etc.
Start with something like 10k, 6k, 3k. After 6 months, you could expand the best performing tier(s) to fine tune the pricing. Let's say the 6k tier produced the best results by far, great, re-randomize the tiers to 8k, 7k, 6k, 5k, 4k and let them run for the rest of the experiment.

If you need help with anything, drop me a DM

After some consideration and taking into account the comments in this thread, I have decided to start an experiment in December with lower prices, as more reliable results will come faster due to a larger number of sold domains. For now, I plan to divide all the domains into 4 groups and set a different price for each group.

Group 1: 999
Group 2: 1999
Group 3: 2999
Group 4: 3999

Then, I'll wait 3-6 months (as the process progresses, it will be clear how many sales occur in each group and how accurate the results are). After that, I will decide whether to increase or decrease the prices for the experiment. For example, if the most profitable option turns out to be 999, it will likely be necessary to run a new experiment with prices like 888, 999, 1588, 1888, or something similar. Conversely, if the price of 3999 proves to be the most advantageous, then it will be possible to test higher prices.

I have tried setting different prices for the domains over the past year, and so far, the most successful prices have been 1999 and 4999, which, after deducting all acquisition and commission costs, yielded about the same amount needed to renew the domains (a scenario that suits me perfectly).

Do you have a link to your portfolio?

No

interesting

I share my experience. I have a very small domain portfolio, and I offer a 25-35% discount twice a year: the first is a summer discount from July 20 to September 10, and the second is a Christmas promo from November 20 to January 10. As a result, I sold 4 domains during these discount periods (out of a total of 6 sales for my 4L.com domains).”

My main goal is to find the price that will work most effectively for a large number of domains with random letter combinations while minimizing the time spent. Purely based on probability, there will always be buyers for these domains (new businesses, funds, individuals, etc) who need such domains in the .com zone, but it's one thing to pay 999 for the needed domain, and another to pay 14,999 or 100,000. I would like to test prices in the 100-200k range, but I don't have the time to wait for the experiment results